{"schema_version":"onlylabs.public_signal.v1","title":"Baseten Writing: Making Kimi K3 Tokenization 18x Faster For Million Token Agentic Workloads","description":"Baseten writing signal with public source context, captured evidence pages, related signals, and category-scoped analysis context.","url":"https://onlylabs.fyi/signals/85638ca4-ea1a-4c7e-93a9-f2a1d7799f8b","json_url":"https://onlylabs.fyi/signals/85638ca4-ea1a-4c7e-93a9-f2a1d7799f8b/signal.json","generated_at":"2026-07-28T19:26:41.201Z","evidence_latest_fetched_at":"2026-07-28T00:02:48.848973+00:00","signal_first_seen_at":"2026-07-28T00:01:46.361107+00:00","org":{"slug":"baseten","name":"Baseten","category":"neocloud","category_label":"Neocloud","dossier_url":"https://onlylabs.fyi/labs/baseten","dossier_json_url":"https://onlylabs.fyi/labs/baseten/dossier.json"},"related_urls":{"signal":"https://onlylabs.fyi/signals/85638ca4-ea1a-4c7e-93a9-f2a1d7799f8b","signal_json":"https://onlylabs.fyi/signals/85638ca4-ea1a-4c7e-93a9-f2a1d7799f8b/signal.json","source":"https://www.baseten.co/blog/making-kimi-k3-tokenization-18x-faster-for-million-token-agentic-workloads/","lab_dossier":"https://onlylabs.fyi/labs/baseten","lab_dossier_json":"https://onlylabs.fyi/labs/baseten/dossier.json","analysis":"https://onlylabs.fyi/analysis/baseten","analysis_json":"https://onlylabs.fyi/analysis/baseten/analysis.json","analysis_evidence_json":"https://onlylabs.fyi/analysis/baseten/evidence.json","category":"https://onlylabs.fyi/neoclouds","category_json":"https://onlylabs.fyi/neoclouds.json","category_feed":"https://onlylabs.fyi/neoclouds/feed.xml","category_signals_json":"https://onlylabs.fyi/signals.json?category=neocloud","topic":"https://onlylabs.fyi/topics/talking","topic_signals_json":"https://onlylabs.fyi/topics/talking/signals.json?category=neocloud","topic_feed":"https://onlylabs.fyi/topics/talking/feed.xml?category=neocloud","data_business":null},"answer_pack":{"answer":"Baseten published Making Kimi K3 Tokenization 18x Faster For Million Token Agentic Workloads. This talking signal gives public context for research themes, product direction, policy, or launch framing. High-signal details: Making Kimi K3 tokenization 18x faster for million-token agentic workloads Kimi K3 is here. Try it now Model performance Making Kimi K3 tokenization 18x faster for.... onlylabs links this event to 1 captured evidence page and 6 related writing signals.","signal_desk":"talking","source_context":{"source_url":"https://www.baseten.co/blog/making-kimi-k3-tokenization-18x-faster-for-million-token-agentic-workloads/","source_host":"baseten.co","occurred_at":"2026-07-27T21:41:19+00:00","first_seen_at":"2026-07-28T00:01:46.361107+00:00","date_source":"sitemap.lastmod","context":null},"context_markers":[{"label":"Lab","value":"Baseten","source":"signal"},{"label":"Signal desk","value":"talking","source":"signal"},{"label":"Source host","value":"baseten.co","source":"source"},{"label":"Model refs","value":"baseten/kimi-k3-tokenizer","source":"model"},{"label":"Watch term","value":"Eval methodology","source":"evidence"},{"label":"Watch term","value":"Model card","source":"model"},{"label":"Watch term","value":"Infrastructure","source":"evidence"},{"label":"Watch term","value":"Safety and alignment","source":"evidence"},{"label":"Watch term","value":"Agents and tool use","source":"evidence"}],"evidence_coverage":{"target_pages":1,"captured_pages":1,"readable_pages":1,"capture_methods":["plain"],"missing_page_urls":[],"failed_page_urls":[],"blocked_page_urls":[],"page_urls":["https://www.baseten.co/blog/making-kimi-k3-tokenization-18x-faster-for-million-token-agentic-workloads/"],"related_signals":6,"has_source_url":true,"latest_page_fetched_at":"2026-07-28T00:02:48.848973+00:00"},"data_business":{"matches":false,"lanes":[],"matched_terms":[],"score":null,"reason":null},"agent_handoff":{"signal_json":"https://onlylabs.fyi/signals/85638ca4-ea1a-4c7e-93a9-f2a1d7799f8b/signal.json","dossier_json":"https://onlylabs.fyi/labs/baseten/dossier.json","analysis_json":"https://onlylabs.fyi/analysis/baseten/analysis.json","analysis_evidence_json":"https://onlylabs.fyi/analysis/baseten/evidence.json","topic_signals_json":"https://onlylabs.fyi/topics/talking/signals.json?category=neocloud","topic_feed":"https://onlylabs.fyi/topics/talking/feed.xml?category=neocloud","category_signals_json":"https://onlylabs.fyi/signals.json?category=neocloud","data_radar_json":null,"opportunities_json":null},"analysis_playbook":{"objective":"Turn public writing and discussion into a readable map of research themes, product framing, policy posture, launch narratives, and market attention.","evidence_focus":["post title","source URL","captured page text","HN traction","linked model or paper references","publication date"],"extraction_questions":["Which themes are labs choosing to explain publicly?","Which posts are attracting outside discussion?","Which writing reframes a recent release, model, hiring wave, or policy stance?","Which posts mention data, evals, infrastructure, safety, or deployment workflows?"],"signal_questions":["What public theme, launch framing, or research direction does this writing signal expose?","Which themes are labs choosing to explain publicly?","Which posts are attracting outside discussion?","Do the 6 related writing signals show a repeated pattern?"],"output_fields":["org","theme","public_framing","traction","evidence_url"],"data_business_relevance":"Data-business lane extraction is scoped to frontier labs; for this category, keep conclusions tied to category-specific strategy, source evidence, and follow-up questions.","required_sources":[{"label":"signal_json","url":"https://onlylabs.fyi/signals/85638ca4-ea1a-4c7e-93a9-f2a1d7799f8b/signal.json","required":true},{"label":"source","url":"https://www.baseten.co/blog/making-kimi-k3-tokenization-18x-faster-for-million-token-agentic-workloads/","required":true},{"label":"dossier_json","url":"https://onlylabs.fyi/labs/baseten/dossier.json","required":true},{"label":"analysis_evidence_json","url":"https://onlylabs.fyi/analysis/baseten/evidence.json","required":true},{"label":"topic_signals_json","url":"https://onlylabs.fyi/topics/talking/signals.json?category=neocloud","required":false},{"label":"data_radar_json","url":null,"required":false}],"expected_output":["one-paragraph source-grounded interpretation","category-specific implication","confidence and missing evidence","recommended next source to inspect"],"prompt_seed":"Using only the linked onlylabs JSON, captured source context, and cited evidence, analyze Baseten's writing signal \"Making Kimi K3 Tokenization 18x Faster For Million Token Agentic Workloads\" for neocloud strategy."},"semantic_triples":[{"subject":"Baseten","predicate":"published","object":"Making Kimi K3 Tokenization 18x Faster For Million Token Agentic Workloads","text":"Baseten published Making Kimi K3 Tokenization 18x Faster For Million Token Agentic Workloads."},{"subject":"Making Kimi K3 Tokenization 18x Faster For Million Token Agentic Workloads","predicate":"is classified as","object":"writing signal","text":"Making Kimi K3 Tokenization 18x Faster For Million Token Agentic Workloads is classified as writing signal."},{"subject":"Making Kimi K3 Tokenization 18x Faster For Million Token Agentic Workloads","predicate":"belongs to","object":"talking desk","text":"Making Kimi K3 Tokenization 18x Faster For Million Token Agentic Workloads belongs to talking desk."},{"subject":"Making Kimi K3 Tokenization 18x Faster For Million Token Agentic Workloads","predicate":"has evidence coverage","object":"1 captured evidence page","text":"Making Kimi K3 Tokenization 18x Faster For Million Token Agentic Workloads has evidence coverage 1 captured evidence page."},{"subject":"Making Kimi K3 Tokenization 18x Faster For Million Token Agentic Workloads","predicate":"has captured page count","object":"1","text":"Making Kimi K3 Tokenization 18x Faster For Million Token Agentic Workloads has captured page count 1."},{"subject":"Making Kimi K3 Tokenization 18x Faster For Million Token Agentic Workloads","predicate":"has readable page count","object":"1","text":"Making Kimi K3 Tokenization 18x Faster For Million Token Agentic Workloads has readable page count 1."},{"subject":"Making Kimi K3 Tokenization 18x Faster For Million Token Agentic Workloads","predicate":"has related signal count","object":"6","text":"Making Kimi K3 Tokenization 18x Faster For Million Token Agentic Workloads has related signal count 6."},{"subject":"Making Kimi K3 Tokenization 18x Faster For Million Token Agentic Workloads","predicate":"has analysis playbook objective","object":"Turn public writing and discussion into a readable map of research themes, product framing, policy posture, launch narratives, and market attention.","text":"Making Kimi K3 Tokenization 18x Faster For Million Token Agentic Workloads has analysis playbook objective Turn public writing and discussion into a readable map of research themes, product framing, policy posture, launch narratives, and market attention.."},{"subject":"Making Kimi K3 Tokenization 18x Faster For Million Token Agentic Workloads","predicate":"has source host","object":"baseten.co","text":"Making Kimi K3 Tokenization 18x Faster For Million Token Agentic Workloads has source host baseten.co."},{"subject":"Making Kimi K3 Tokenization 18x Faster For Million Token Agentic Workloads","predicate":"has lab","object":"Baseten","text":"Making Kimi K3 Tokenization 18x Faster For Million Token Agentic Workloads has lab Baseten."},{"subject":"Making Kimi K3 Tokenization 18x Faster For Million Token Agentic Workloads","predicate":"has signal desk","object":"talking","text":"Making Kimi K3 Tokenization 18x Faster For Million Token Agentic Workloads has signal desk talking."},{"subject":"Making Kimi K3 Tokenization 18x Faster For Million Token Agentic Workloads","predicate":"has source host","object":"baseten.co","text":"Making Kimi K3 Tokenization 18x Faster For Million Token Agentic Workloads has source host baseten.co."},{"subject":"Making Kimi K3 Tokenization 18x Faster For Million Token Agentic Workloads","predicate":"has model refs","object":"baseten/kimi-k3-tokenizer","text":"Making Kimi K3 Tokenization 18x Faster For Million Token Agentic Workloads has model refs baseten/kimi-k3-tokenizer."},{"subject":"Making Kimi K3 Tokenization 18x Faster For Million Token Agentic Workloads","predicate":"has watch term","object":"Eval methodology","text":"Making Kimi K3 Tokenization 18x Faster For Million Token Agentic Workloads has watch term Eval methodology."},{"subject":"Making Kimi K3 Tokenization 18x Faster For Million Token Agentic Workloads","predicate":"has watch term","object":"Model card","text":"Making Kimi K3 Tokenization 18x Faster For Million Token Agentic Workloads has watch term Model card."},{"subject":"Making Kimi K3 Tokenization 18x Faster For Million Token Agentic Workloads","predicate":"has watch term","object":"Infrastructure","text":"Making Kimi K3 Tokenization 18x Faster For Million Token Agentic Workloads has watch term Infrastructure."},{"subject":"Making Kimi K3 Tokenization 18x Faster For Million Token Agentic Workloads","predicate":"has watch term","object":"Safety and alignment","text":"Making Kimi K3 Tokenization 18x Faster For Million Token Agentic Workloads has watch term Safety and alignment."}]},"intelligence":{"signal_desk":"talking","answer":"Baseten published Making Kimi K3 Tokenization 18x Faster For Million Token Agentic Workloads. This talking signal gives public context for research themes, product direction, policy, or launch framing. High-signal details: Making Kimi K3 tokenization 18x faster for million-token agentic workloads Kimi K3 is here. Try it now Model performance Making Kimi K3 tokenization 18x faster for.... onlylabs links this event to 1 captured evidence page and 6 related writing signals.","semantic_triples":[{"subject":"Baseten","predicate":"published","object":"Making Kimi K3 Tokenization 18x Faster For Million Token Agentic Workloads","text":"Baseten published Making Kimi K3 Tokenization 18x Faster For Million Token Agentic Workloads."},{"subject":"Making Kimi K3 Tokenization 18x Faster For Million Token Agentic Workloads","predicate":"is classified as","object":"writing signal","text":"Making Kimi K3 Tokenization 18x Faster For Million Token Agentic Workloads is classified as writing signal."},{"subject":"Making Kimi K3 Tokenization 18x Faster For Million Token Agentic Workloads","predicate":"belongs to","object":"talking desk","text":"Making Kimi K3 Tokenization 18x Faster For Million Token Agentic Workloads belongs to talking desk."},{"subject":"Making Kimi K3 Tokenization 18x Faster For Million Token Agentic Workloads","predicate":"has evidence coverage","object":"1 captured evidence page","text":"Making Kimi K3 Tokenization 18x Faster For Million Token Agentic Workloads has evidence coverage 1 captured evidence page."}]},"signal":{"id":"85638ca4-ea1a-4c7e-93a9-f2a1d7799f8b","url":"https://onlylabs.fyi/signals/85638ca4-ea1a-4c7e-93a9-f2a1d7799f8b","json_url":"https://onlylabs.fyi/signals/85638ca4-ea1a-4c7e-93a9-f2a1d7799f8b/signal.json","source_url":"https://www.baseten.co/blog/making-kimi-k3-tokenization-18x-faster-for-million-token-agentic-workloads/","title":"Making Kimi K3 Tokenization 18x Faster For Million Token Agentic Workloads","summary":"Baseten published a writing signal. onlylabs watches public writing for research themes, product direction, and model-launch context.","context":null,"kind":{"key":"post_published","label":"Writing"},"org":{"slug":"baseten","name":"Baseten","category":"neocloud"},"occurred_at":"2026-07-27T21:41:19+00:00","first_seen_at":"2026-07-28T00:01:46.361107+00:00","date_source":"sitemap.lastmod","evidence_coverage":{"target_pages":1,"captured_pages":1,"readable_pages":1,"capture_methods":["plain"],"missing_page_urls":[],"failed_page_urls":[],"blocked_page_urls":[],"page_urls":["https://www.baseten.co/blog/making-kimi-k3-tokenization-18x-faster-for-million-token-agentic-workloads/"]},"facets":{},"traction":{"github_stars":null,"hn_points":null,"hn_comments":null,"hn_story_id":null,"hf_downloads":null,"hf_likes":null},"data_radar":null},"primary_evidence_page":{"is_primary":true,"source_match":true,"url":"https://www.baseten.co/blog/making-kimi-k3-tokenization-18x-faster-for-million-token-agentic-workloads/","final_url":"https://www.baseten.co/blog/making-kimi-k3-tokenization-18x-faster-for-million-token-agentic-workloads/","title":"Making Kimi K3 Tokenization 18x Faster For Million Token Agentic Workloads","http_status":200,"content_type":"text/html; charset=utf-8","capture_method":"plain","fetched_at":"2026-07-28T00:02:48.848973+00:00","bytes":353423,"raw_path":"2db7f0ab404434ad6f7a58b5e5bd5303f16db1bcbc9822345716574afc6d8baf.html","content_hash":"775025762e72d13a888580282ad9a8f2966ea977aad366fc5a8016b96e7ed2f3","excerpt_chars":1200,"truncated":true,"excerpt":"Making Kimi K3 tokenization 18x faster for million-token agentic workloads Kimi K3 is here. Try it now Model performance Making Kimi K3 tokenization 18x faster for million-token agentic workloads We built the Baseten Tokenizer (Basetenkenizer) to optimize tokenization for long input sequences, starting with Kimi K3. Authors Michael Feil Last updated July 27, 2026 Share For years, inference engineers have been able to disregard tokenization time as negligible. Short input sequences take a couple of milliseconds to tokenize, a tiny fraction of the time required for prefill and decode. This has changed. Open frontier models like Kimi K3 support input sequences of up to one million tokens. These long input sequences are increasingly common in agentic workloads, where an agent loop repeatedly appends tool results, observations, retrieved documents, intermediate state, reasoning traces, and updated instructions back into the next request. ✕ Input sequence lengths and prefix cache hit rates increase quickly in agentic loops We built the Baseten Tokenizer (Basetenkenizer) to optimize tokenization for long input sequences, starting with Kimi K3. Within the Baseten Inference Stack, we..."},"evidence_pages":[],"related_signals":[{"id":"582a5903-83a9-4773-af4a-4cc9707a31d6","url":"https://onlylabs.fyi/signals/582a5903-83a9-4773-af4a-4cc9707a31d6","source_url":"https://www.baseten.co/resources/changelog/kimi-k3-available-on-baseten/","title":"Kimi K3 Available On Baseten","context":null,"kind":{"key":"post_published","label":"Writing"},"org":{"slug":"baseten","name":"Baseten","category":"neocloud"},"occurred_at":"2026-07-27T15:17:02+00:00","first_seen_at":"2026-07-27T16:00:51.500481+00:00","date_source":"sitemap.lastmod"},{"id":"a3f0d254-cb58-4c4f-a9da-a5493000afd9","url":"https://onlylabs.fyi/signals/a3f0d254-cb58-4c4f-a9da-a5493000afd9","source_url":"https://www.baseten.co/blog/how-to-build-a-day-zero-api-for-kimi-k3/","title":"How To Build A Day Zero Api For Kimi K3","context":null,"kind":{"key":"post_published","label":"Writing"},"org":{"slug":"baseten","name":"Baseten","category":"neocloud"},"occurred_at":"2026-07-27T15:16:02+00:00","first_seen_at":"2026-07-27T16:00:51.500481+00:00","date_source":"sitemap.lastmod"},{"id":"eeba17cb-81d1-4b1d-8947-d1ea51e8ecd3","url":"https://onlylabs.fyi/signals/eeba17cb-81d1-4b1d-8947-d1ea51e8ecd3","source_url":"https://www.baseten.co/blog/how-we-built-the-new-fastest-api-for-glm-52/","title":"How We Built The New Fastest Api For Glm 52","context":null,"kind":{"key":"post_published","label":"Writing"},"org":{"slug":"baseten","name":"Baseten","category":"neocloud"},"occurred_at":"2026-07-26T02:47:03+00:00","first_seen_at":"2026-07-26T04:00:50.760097+00:00","date_source":"sitemap.lastmod"},{"id":"1761f1d3-f0a1-4265-9b92-b80878abe063","url":"https://onlylabs.fyi/signals/1761f1d3-f0a1-4265-9b92-b80878abe063","source_url":"https://www.baseten.co/resources/changelog/health-check-metrics/","title":"Health Check Metrics","context":null,"kind":{"key":"post_published","label":"Writing"},"org":{"slug":"baseten","name":"Baseten","category":"neocloud"},"occurred_at":"2026-07-24T05:01:54+00:00","first_seen_at":"2026-07-24T08:00:52.178883+00:00","date_source":"sitemap.lastmod"},{"id":"45fb4517-c6cb-4bab-9b99-a7189c15de4a","url":"https://onlylabs.fyi/signals/45fb4517-c6cb-4bab-9b99-a7189c15de4a","source_url":"https://www.baseten.co/resources/changelog/glm-52-fast-available-on-baseten/","title":"Glm 52 Fast Available On Baseten","context":null,"kind":{"key":"post_published","label":"Writing"},"org":{"slug":"baseten","name":"Baseten","category":"neocloud"},"occurred_at":"2026-07-23T19:42:33+00:00","first_seen_at":"2026-07-23T20:00:52.856312+00:00","date_source":"sitemap.lastmod"},{"id":"a7cdba8a-28c3-42b4-a7a3-ace8f5110d5c","url":"https://onlylabs.fyi/signals/a7cdba8a-28c3-42b4-a7a3-ace8f5110d5c","source_url":"https://www.baseten.co/resources/changelog/api-key-management-keys/","title":"Api Key Management Keys","context":null,"kind":{"key":"post_published","label":"Writing"},"org":{"slug":"baseten","name":"Baseten","category":"neocloud"},"occurred_at":"2026-07-23T19:23:54+00:00","first_seen_at":"2026-07-23T20:00:52.856312+00:00","date_source":"sitemap.lastmod"}]}